# The Carbohydrate–Insulin Model of Obesity: research update

Dated 5 October 2026. Local prepublication addendum; archived grade unchanged.

Diet interventions can change insulin, expenditure and intake, but several predictions of the broad causal model fail in particular controlled settings. Adaptation, diet order, food properties and expenditure measurement remain material sources of disagreement.

## Research timeline

Electronic year is used where recorded; issue year is retained in each source card. The chronology is selected, not exhaustive, and does not establish priority or causation.

| Year | Primary study | Design and finding | Inference limit |
|---|---|---|---|
| 1999 | [High glycemic index foods, overeating, and obesity.](https://pubmed.ncbi.nlm.nih.gov/10049982/) | Acute crossover in adolescents. High-glycemic-index meals increased subsequent voluntary intake in the tested setting. | Acute meal effects do not establish chronic obesity initiation. |
| 2015 | [Calorie for Calorie, Dietary Fat Restriction Results in More Body Fat Loss than Carbohydrate Restriction in People with Obesity.](https://pubmed.ncbi.nlm.nih.gov/26278052/) | Short metabolic-ward randomized crossover. Fat restriction produced more calculated body-fat loss than carbohydrate restriction at matched calories. | Six-day comparisons do not settle all longer adaptation predictions. |
| 2016 | [Energy expenditure and body composition changes after an isocaloric ketogenic diet in overweight and obese men.](https://pubmed.ncbi.nlm.nih.gov/27385608/) | Sequential isocaloric ketogenic feeding. Expenditure increased modestly while body-fat loss did not accelerate. | Sequence, energy imbalance and measurement precision bound inference. |
| 2018 | [Effect of Low-Fat vs Low-Carbohydrate Diet on 12-Month Weight Loss in Overweight Adults and the Association With Genotype Pattern or Insulin Secretion: The DIETFITS Randomized Clinical Trial.](https://pubmed.ncbi.nlm.nih.gov/29466592/) | DIETFITS randomized trial. Weight loss did not significantly differ; baseline insulin secretion did not select the better diet. | Free-living diet advice does not isolate every proposed mediator. |
| 2018 | [Effects of a low carbohydrate diet on energy expenditure during weight loss maintenance: randomized trial.](https://pubmed.ncbi.nlm.nih.gov/30429127/) | Weight-maintenance randomized feeding. Lower carbohydrate intake increased reported expenditure after weight loss. | Outpatient adherence and expenditure-estimation assumptions need scrutiny; correction notice retained. |
| 2021 | [Effect of a plant-based, low-fat diet versus an animal-based, ketogenic diet on ad libitum energy intake.](https://pubmed.ncbi.nlm.nih.gov/33479499/) | Ad libitum inpatient crossover. The high-glycemic-load low-fat diet produced lower intake than the ketogenic comparator. | Food properties and short crossover periods limit attribution to a single pathway. |
| 2024 | [Diet order significantly affects energy balance for diets varying in macronutrients but not ultraprocessing in crossover studies without a washout period.](https://pubmed.ncbi.nlm.nih.gov/39163976/) | Secondary analyses of two crossover trials. Diet order affected the macronutrient study but not the ultra-processing study. | Reuses original cohorts; this is a sensitivity analysis, not an independent replication. |
| 2025 | [Testing the carbohydrate-insulin model: Short-term metabolic responses to consumption of meals with varying glycemic index in healthy adults.](https://pubmed.ncbi.nlm.nih.gov/40043690/) | Randomized acute glycemic-index meals. Glucose/insulin responses differed, while hunger and individual intake relations gave little model support. | Acute healthy-adult responses do not test weeks-long adaptation in common obesity. |

## Research gaps

### G1: Do adaptation and order explain conflicting intake effects?

Evidence anchors: [P27385608](https://pubmed.ncbi.nlm.nih.gov/27385608/), [P33479499](https://pubmed.ncbi.nlm.nih.gov/33479499/), [P39163976](https://pubmed.ncbi.nlm.nih.gov/39163976/), [P40043690](https://pubmed.ncbi.nlm.nih.gov/40043690/).

Competing explanation: Carryover, gut adaptation, fibre or food mass may dominate short crossover contrasts.

Discriminating test and controls: Longer randomized feeding with washout or parallel arms, prespecified adaptation windows and intake endpoints; match food properties.

### G2: Is an expenditure advantage reproducible across methods?

Evidence anchors: [P27385608](https://pubmed.ncbi.nlm.nih.gov/27385608/), [P30429127](https://pubmed.ncbi.nlm.nih.gov/30429127/).

Competing explanation: Doubly labelled water assumptions, energy imbalance and adherence may shift estimated effects.

Discriminating test and controls: Concurrent chamber calorimetry, isotope methods and measured food intake; prespecify correction models, baseline and energy-balance checks.

### G3: Does insulin–glucagon change mediate hunger or adiposity?

Evidence anchors: [P10049982](https://pubmed.ncbi.nlm.nih.gov/10049982/), [P26278052](https://pubmed.ncbi.nlm.nih.gov/26278052/), [P40043690](https://pubmed.ncbi.nlm.nih.gov/40043690/).

Competing explanation: Insulin may respond to intake, while sensory, intestinal or reward mechanisms affect hunger.

Discriminating test and controls: Safe controlled mechanistic studies and longitudinal mediation within diet randomization; control glucose, protein, fibre and energy density.

### G4: Which phenotype benefits, over what time scale?

Evidence anchors: [P30429127](https://pubmed.ncbi.nlm.nih.gov/30429127/), [P29466592](https://pubmed.ncbi.nlm.nih.gov/29466592/), [P33479499](https://pubmed.ncbi.nlm.nih.gov/33479499/).

Competing explanation: Weight-loss maintenance, untreated obesity and acute healthy-adult responses may have different effect modifiers.

Discriminating test and controls: Prespecified insulin-secretion and disease-stage strata in independently replicated longer trials; compare interactions rather than post hoc responder subsets.

## Methods and access

Case-specific Consensus records assisted gap discovery, chronology and targeted paper questions. Shared template wording about an amyloid-specific Deep follow-up is not evidence of a Deep run for this topic; the dated source records establish the actual use and access limits. Primary identifiers, selected abstract passages, correction notices and available primary-body passages were cross-checked. Generated absence cells, duplicate milestones and unsupported access claims were rejected.

This focused addendum does not establish 50–70 complete primary Methods/Results examinations per review. Baseline reading receipts and scores retain their original scope.

Checked timeline sources: 8; selected full-text checks: 0; abstract-excerpt checks: 8; metadata-only historical anchors: 0. Complete papers examined in this update: 0.

Historical scorecards declare schema 5.4. This enrichment uses HSEF v5.3 causal distinctions without retrospectively relabelling or rescoring those files.

Targeted paper-check scope: The targeted Consensus answer reported abstract-only access. Primary abstract and metadata were checked; no independently readable body was verified in this update.

Editorial correction/limit: Consensus attributed some milestones to later commentary or review papers. The chronology uses primary identifiers and keeps secondary cohort analyses explicit.